FinanceBenchmark: Financial AI Leaderboard 2026

FinanceBenchmark provides independent rankings and a composite score for leading Artificial Intelligence models, including Anthropic Claude, OpenAI GPT, and Google Gemini, on complex financial reasoning and forensic accounting tasks. Enable JavaScript to view the interactive, continuously updated leaderboard.

2026-09-10MODELDeepSeek-V4.1-Flash released with extreme agent memory efficiency and evaluated across VALS • 2026-09-03MODELGPT-6 Astra released by OpenAI - 71.7% on EMB, 63.3% on Tax Agent Bench • 2026-09-02MODELGoogle Gemini 3.8 Flash takes #1 rank on Finance Agent (v2) with 61.4% and 72.2% on EMB • 2026-09-02MODELMeta releases Muse Spark 1.3 & 1.3 Max - 59.0%-60.0% on Finance Agent (v2) • 2026-09-01MODELClaude Fable 5.1 leads Tax Agent Bench (77.6%) and EMB (76.7%) with 75% cheaper cache reads • 2026-09-11BENCHTax Agent Bench and FAB v2 benchmarks updated with latest frontier model rankings • 2026-09-15UPDATEPlatform updated: 43 models tracked across 41 independent benchmarks • 2026-09-10MODELDeepSeek-V4.1-Flash released with extreme agent memory efficiency and evaluated across VALS • 2026-09-03MODELGPT-6 Astra released by OpenAI - 71.7% on EMB, 63.3% on Tax Agent Bench • 2026-09-02MODELGoogle Gemini 3.8 Flash takes #1 rank on Finance Agent (v2) with 61.4% and 72.2% on EMB • 2026-09-02MODELMeta releases Muse Spark 1.3 & 1.3 Max - 59.0%-60.0% on Finance Agent (v2) • 2026-09-01MODELClaude Fable 5.1 leads Tax Agent Bench (77.6%) and EMB (76.7%) with 75% cheaper cache reads • 2026-09-11BENCHTax Agent Bench and FAB v2 benchmarks updated with latest frontier model rankings • 2026-09-15UPDATEPlatform updated: 43 models tracked across 41 independent benchmarks •

Tier 3agentTasksforensic

Harvey's Legal Agent

Enterprise legal agent evaluation covering contract analysis, clause drafting, and risk assessment.

Variable tasks

22 models evaluated

Updated 2026

View Source →Submit your score Methodology

25.4%

Top Score

Top model

Muse Spark 1.2

Model Rankings

financebenchmark.ai · Harvey Legal

10.8%

🥇

AnthropicClaude Opus 56.7%

🥈

AnthropicClaude Sonnet 55.0%

🥉

OpenAIGPT-5.6 Sol2.5%

4

DeepSeekDeepSeek V4 FlashOPEN8.3%

5

xAIGrok 4.615.8%

6

Moonshot AIKimi K3OPEN10.8%

7

AnthropicClaude Fable 511.3%

8

DeepSeekDeepSeek V4 Pro (0813)OPEN7.5%

9

GoogleGemini 3.5 Flash Lite0.0%

10

GoogleGemini 3.6 Flash3.3%

11

Zhipu AIGLM 5.2OPEN7.1%

12

OpenAIGPT-5.6 Luna1.3%

13

OpenAIGPT-5.6 Terra0.4%

14

xAIGrok 4.512.9%

15

Thinking MachinesInklingOPEN2.1%

16

Thinking MachinesInkling SmallOPEN1.7%

17

Ant GroupLing 3.0 FlashOPEN1.3%

18

MetaMuse Spark 1.120.0%

19

MetaMuse Spark 1.225.4%

20

NVIDIANemotron 3 UltraOPEN0.4%

21

AlibabaQwen 3.7 Plus0.0%

22

AlibabaQwen 3.8 MaxOPEN10.4%

Full Results Table

#	Model	Overall	Source	Date
🥇	Muse Spark 1.2 Meta	25.4	VALS (2026)	2026-08
🥈	Muse Spark 1.1 Meta	20.0	VALS (2026)	2026-07
🥉	Grok 4.6 xAI	15.8	VALS (2026)	2026-08
4	Grok 4.5 xAI	12.9	VALS (2026)	2026-07
5	Claude Fable 5 Anthropic	11.3	VALS (2026)	2026-06
6	Kimi K3OPEN Moonshot AI	10.8	VALS (2026)	2026-07
7	Qwen 3.8 MaxOPEN Alibaba	10.4	VALS (2026)	2026-08
8	DeepSeek V4 FlashOPEN DeepSeek	8.3	VALS (2026)	2026-07
9	DeepSeek V4 Pro (0813)OPEN DeepSeek	7.5	VALS (2026)	2026-08
10	GLM 5.2OPEN Zhipu AI	7.1	VALS (2026)	2026-06
11	Claude Opus 5 Anthropic	6.7	VALS (2026)	2026-07
12	Claude Sonnet 5 Anthropic	5.0	VALS (2026)	2026-06
13	Gemini 3.6 Flash Google	3.3	VALS (2026)	2026-07
14	GPT-5.6 Sol OpenAI	2.5	VALS (2026)	2026-07
15	InklingOPEN Thinking Machines	2.1	VALS (2026)	2026-07
16	Inkling SmallOPEN Thinking Machines	1.7	VALS (2026)	2026-07
17	GPT-5.6 Luna OpenAI	1.3	VALS (2026)	2026-07
18	Ling 3.0 FlashOPEN Ant Group	1.3	VALS (2026)	2026-07
19	GPT-5.6 Terra OpenAI	0.4	VALS (2026)	2026-07
20	Nemotron 3 UltraOPEN NVIDIA	0.4	VALS (2026)	2026-06
21	Gemini 3.5 Flash Lite Google	0.0	VALS (2026)	2026-07
22	Qwen 3.7 Plus Alibaba	0.0	VALS (2026)	2026-06